INDEX

Explanations

that provide

The neuron activates on numeric tokens—integers, decimals, and other number‐like values.

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

Negative Logits

áciu

-1.39

 туда

-1.24

 aint

-1.24

uktion

-1.22

rů

-1.22

有问题

-1.22

 такого

-1.20

 trovare

-1.20

 antichi

-1.17

锱

-1.16

POSITIVE LOGITS

to

1.58

1.32

⎟

1.30

 повече

1.29

嘢

1.27

Е

1.25

 sesama

1.23

たぶん

1.22

本当

1.20

Activations Density 0.060%